Writing.

Thoughts, technical deep dives, and perspectives on engineering, architecture, and design.

Articles & Publications

Writing.

Thoughts, technical deep dives, and perspectives on engineering, architecture, and design.

Inside Jev and System One Models: Why Giving Up Text Generation Makes AI Fast and Type-Safe

September 26, 2026

Inside Jev and System One Models: Why Giving Up Text Generation Makes AI Fast and Type-Safe

A technical breakdown of TypeSafe AI's Jev model, exploring why parallel sampling and RLCD replace autoregressive text generation for fast, schema-guaranteed decisions.

2 min read

Read Article

Arch Linux vs. The Wild West: How the Latest AUR Malware Attack Went Down

Sep 13, 2026·2 min read

Arch Linux vs. The Wild West: How the Latest AUR Malware Attack Went Down

Arch users, stop blindly running makepkg -si. Here is how the latest AUR malware attack worked and how to keep your system safe.

#arch-linux#aur#bash

Generative AI Demystified: It's Just Fancy Autocomplete (And That's Okay)

Sep 13, 2026·2 min read

Generative AI Demystified: It's Just Fancy Autocomplete (And That's Okay)

Demystifying Generative AI without the hype. Learn how LLMs actually work under the hood using probability, tokens, and a bit of math.

#beginners#generative-ai#LLM

The Anatomy of a High-Quality Prompt: An Engineering Approach to LLM Steering

Sep 13, 2026·2 min read

The Anatomy of a High-Quality Prompt: An Engineering Approach to LLM Steering

Learn the mechanics of prompt engineering. Understand how LLMs process inputs and how to structure prompts using system instructions, context, and constraints.

Demystifying the 4 Pillars of OOP in Java (Without the Corporate Jargon)

Sep 13, 2026·2 min read

Demystifying the 4 Pillars of OOP in Java (Without the Corporate Jargon)

Learn the four pillars of OOP in Java with fun analogies, clean code, and zero boring textbook definitions. Perfect for beginners!

#Backend#Java#oop

Sep 13, 2026·2 min read

Beyond the Desktop Flame Wars

A deep architectural comparison between Linux and Windows, exploring kernel design, process scheduling, memory management, and enterprise server realities.

#DevOps#Kernel#Linux

Why the US Government Banned Claude (And What It Teaches Us About AI Data Privacy)

Sep 13, 2026·2 min read

Why the US Government Banned Claude (And What It Teaches Us About AI Data Privacy)

The US House banned Claude for staff. Here is the technical breakdown of why public LLMs scare security teams and how to build secure AI pipelines.

#AI Security#Anthropic#Claude

How to Copy Homework Worth $100 Million: Anthropic vs. Alibaba

Sep 13, 2026·2 min read

How to Copy Homework Worth $100 Million: Anthropic vs. Alibaba

Anthropic accuses Alibaba of secretly copying Claude's brain. Here is the hilarious and technical breakdown of AI model distillation and how tech giants copy homework.

#AI#Alibaba#Anthropic